Abstract

Background Interventions involving both exercise and dietary modification are effective in reducing steatosis in nonalcoholic fatty liver disease (NAFLD). However, exercise alone may reduce liver fat and is known to have other positive effects on health. The primary aim of this study was to systematically review the effect of exercise alone without dietary intervention on NAFLD and to examine correlations across changes in liver fat and metabolic markers during exercise. Methods Relevant online databases were searched from earliest records to May 2020 by two researchers. Studies were included where the trial was a randomized controlled trial, participants were adults, exercise intervention was longer than 4 weeks, no dietary intervention occurred, and the effect of the intervention on liver fat was quantified via magnetic resonance imaging/proton magnetic resonance spectroscopy. Results Of 21 597 studies retrieved, 16 were included involving 706 participants. Exercise was found to have a beneficial effect on liver fat without dietary modification (-2.4%, -3.13 to -1.66) (mean, 95% CI). Pearson correlation showed significant relationships between change in liver fat and change in weight (r = 0.67,P= .007), liver enzymes aspartate aminotransferase (r = 0.76,P= .002) and alanine aminotransferase (r = 0.91,P < .001), and cardiorespiratory fitness VO(2)peak (peak volume oxygen consumption) (r = -0.88,P= .004). By multivariate regression, change in weight and change in VO(2)peak significantly contributed to change in liver fat (R-2= 0.84,P= .01). Conclusions This systematic review found that exercise without dietary intervention improves liver fat and that clinical markers may be useful proxies for quantifying liver fat changes.